Clinical and prognostic implications of KRT19 expression in pancreatic ductal adenocarcinoma

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is an extremely aggressive cancer that is usually detected late and responds poorly to current therapies. PDAC is the most common pancreatic neoplasm and accounts for more than 90% of all pancreatic cancer cases. Even with advances in available treatments, the 5-year survival rate remains below 10% (1-3).

Furthermore, at the time of diagnosis, nearly 80–85% of PDAC cases are identified in an advanced, metastatic, or unresectable state. Even for patients who are fortunate enough to be diagnosed at a resectable stage and undergo proper surgical resection, the prognosis remains unfavorable because of the high recurrence rate (4,5).

PDAC is frequently diagnosed at an advanced stage and is characterized by a high recurrence rate and an aggressive clinical course. Therefore, biomarkers that can assist in early detection and prognosis prediction are urgently needed (6).

Keratin 19 (KRT19) is a type I keratin protein of approximately 40 kDa encoded by the KRT19 gene (7,8). It is normally expressed in the epithelial tissues and is generally absent in the peripheral blood of healthy individuals. However, with the advancement of sensitive laboratory methods, small amounts of circulating KRT19 protein have been detected in patients with epithelial-derived malignancies. Moreover, aberrant KRT19 overexpression has been reported in various cancers, including breast cancer, hepatocellular carcinoma, pancreatic cancer, and renal cell neoplasm (8-11).

In this regard, KRT19 may play an important role as a candidate biomarker for early cancer detection and prognostic evaluation, particularly in PDAC (5,12-14).

This study investigated the prognostic role of KRT19 expression in PDAC using publicly available datasets and bioinformatics analyses. We present this article in accordance with the REMARK reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2629/rc).

Methods

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Gene expression data source

We retrieved pancreatic cancer gene expression profiles from the public National Center for Biotechnology Information Gene Expression Omnibus (GEO; RRID:SCR_005012) repository (National Institutes of Health, Bethesda, MD, USA). Two datasets, GSE15471 and GSE22780, produced on the GPL570 platform (Affymetrix Human Genome U133 Plus 2.0 Array) were selected for this study. The Affymetrix probe set corresponding to KRT19 (201650_at) on the GPL570 platform was used for expression analysis. GSE15471 included data from 39 pancreatic carcinoma tissues and 36 normal pancreatic tissues, whereas GSE22780 included data from eight cancer samples and eight normal controls. A total of 47 tumor samples and 44 normal pancreatic samples were analyzed. The raw microarray data were processed in R (version 4.2.3; RRID:SCR_001905) using the Affy package (version 1.76.0). To integrate the two GEO cohorts and minimize technical batch variations, raw data from both datasets were pooled and subjected to a robust multi-array average (RMA) procedure, which includes background correction, quantile normalization, and log₂-transformation.

Clinical data source

Gene expression RNA sequencing data (dataset ID: TCGA-PAAD.star_fpkm.tsv), clinicopathological characteristics (dataset ID: TCGA-PAAD.clinical.tsv) and survival outcomes (dataset ID: TCGA-PAAD.survival.tsv) of patients with PDAC were obtained from the University of California Santa Cruz (UCSC) Xena Browser (http://xenabrowser.net/; RRID:SCR_018938) for The Cancer Genome Atlas Pancreatic Adenocarcinoma (TCGA-PAAD; RRID:SCR_003193) cohort. To normalize the data distribution prior to downstream analysis, the raw Fragments Per Kilobase of transcript per Million mapped reads (FPKM) values were log2-transformed [log₂(FPKM + 1)]. A total of 177 patients with PDAC were analyzed after excluding those with insufficient clinical data. Clinical variables included age, tumor grade, American Joint Committee on Cancer (AJCC) Cancer stage, diabetes, history of chronic pancreatitis, alcohol consumption, smoking history, TP53 mutations, and KRAS mutations. The mean expression of KRT19 was compared among clinical parameters. This study utilized de-identified, publicly accessible datasets.

Survival analysis of KRT19 expression level and association with clinicopathological features

The primary outcome of this study was overall survival of patients with PDAC. The patients were divided into high and low KRT19 expression groups according to the median expression level. Among the total of 177 patients, one case lacked survival information, and thus 176 patients were included in the survival analysis.

Secondary outcomes included the association of KRT19 expression with clinicopathological features such as tumor grade, AJCC stage, and TP53/KRAS mutational status.

Associations between KRT19 expression and clinical variables such as chronic pancreatitis, smoking, and diabetes were also explored.

Statistical analysis

Gene expression levels in normal and tumor tissues were compared using Student’s t-test. Statistical significance was set at P<0.05. One-way analysis of variance (ANOVA) was performed to compare gene expression levels across more than two groups. The results were considered statistically significant at P<0.05. For survival analysis, patients were stratified into high and low expression groups according to the median KRT19 expression level, which served as the cutoff value. Kaplan-Meier survival curves were generated using the log-rank test, and differences between groups were evaluated. Statistical significance was set at P<0.05.

To evaluate this association, mutation data for TP53 and KRAS were integrated with KRT19 expression. Differences in KRT19 expression between the mutant and wild-type groups were evaluated using a t-test. The results were considered statistically significant at P<0.05.

The Cox proportional hazards regression model was used for both the univariate and multivariate survival analyses. Variables that showed significance in the univariate test (P<0.05) were subsequently included in the multivariate analysis. All statistical analyses and visualizations were performed using R software (version 4.2.3) with standard packages including survival, survminer, and ggplot2. Statistical significance was set at a threshold of P<0.05.

Results

KRT19 expression is significantly upregulated in pancreatic cancer tissues

To investigate the expression pattern of KRT19 in PDAC, gene expression profiles were analyzed using GEO datasets (GSE15471 and GSE22780). A total of 44 normal pancreatic tissue samples and 47 tumor tissue samples were included. The median KRT19 expression level was 6.636 in normal tissues (N=44) and 10.538 in tumor tissues (N=47). Analysis demonstrated that KRT19 levels were markedly higher in tumor tissues than in normal pancreatic tissues (P<0.001; Figure 1).

Figure 1 Comparison of KRT19 expression levels between normal pancreatic tissues and PDAC tissues. KRT19 expression was analyzed using integrated GEO datasets (GSE15471 and GSE22780). The analysis included 44 normal pancreatic tissue samples and 47 tumor tissue samples. The middle line in the box plot represents the median value. Statistical significance was determined using Student’s t-test. KRT19 expression was significantly higher in tumor tissues (median: 10.538) compared to normal tissues (median: 6.636) (P<0.001). GEO, Gene Expression Omnibus; KRT19, keratin 19; PDAC, pancreatic ductal adenocarcinoma.

Relationship between KRT19 expression and clinical characteristics

The relationship between KRT19 expression and the clinicopathological characteristics of patients with PDAC was further investigated. KRT19 expression did not differ significantly with age, diabetes, smoking, or alcohol consumption. However, a significant association was observed between KRT19 overexpression, and AJCC stage, tumor grade, TP53 mutation and KRAS mutation. Higher KRT19 expression was observed in tumors classified as AJCC stage ≥2b than in those classified as stage ≤2a (9.089 vs. 8.654 FPKM; P<0.001; Table 1). The median KRT19 expression level was also markedly upregulated in higher-grade tumors (grade 1, 8.233; grade 2, 9.328; and grade 3, 9.504; P<0.001; Figure 2). These findings indicated that KRT19 overexpression was positively correlated with advanced disease features in PDAC. Furthermore, KRT19 expression was positively correlated with history of chronic pancreatitis (mean: 9.951 vs. 8.821, P=0.03).

Figure 2 Association between KRT19 expression and histological tumor grade in PDAC. KRT19 expression levels were evaluated across different histological grades (grade 1, n=31; grade 2, n=94, grade 3, n=52) using the TCGA-PAAD cohort (n=177). One-way analysis of variance was used to compare expression levels between groups. Median KRT19 expression level was significantly higher in advanced grade tumors (grade 1, 8.233; grade 2, 9.328; grade 3, 9.504; P<0.001). KRT19, keratin 19; PDAC, pancreatic ductal adenocarcinoma; TCGA-PAAD, The Cancer Genome Atlas-Pancreatic Adenocarcinoma.

Patients with higher KRT19 expression had worse overall survival

Kaplan-Meier survival analysis was performed using TCGA Pancreatic Cancer cohort. The patients were stratified into high-and low-expression groups according to the median KRT19 expression level. Patients in the high-expression group showed significantly poorer overall survival than those in the low-expression group [hazards ratio (HR) =1.795, 95% confidence interval (CI): 1.182–2.727, P=0.005; Figure 3].

Figure 3 Kaplan-Meier survival curves for overall survival based on KRT19 expression in PDAC. Patients from the TCGA-PAAD cohort (n=176) were stratified into high and low expression groups according to the median KRT19 expression level. The high-expression group showed significantly poor overall survival compared to the low-expression group. Statistical significance was evaluated using the log-rank test (HR =1.795, 95% CI: 1.182–2.727, P=0.005). CI, confidence interval; HR, hazards ratio; KRT19, keratin 19; PDAC, pancreatic ductal adenocarcinoma; TCGA-PAAD, The Cancer Genome Atlas-Pancreatic Adenocarcinoma.

KRT19 expression level is positively correlated with TP53 and KRAS mutations and predicts poor prognosis

The association of KRT19 expression with TP53 and KRAS mutations was also analyzed. Patients harboring TP53 mutations demonstrated significantly higher KRT19 expression than wild-type patients (P<0.001; Figure 4A) and had worse overall survival (HR =1.667, P=0.02; Figure 4B). Similarly, KRT19 expression was upregulated in KRAS-mutated tumors (P<0.001; Figure 4C), with poorer survival outcomes than in wild-type tumors (HR =1.807, P=0.02; Figure 4D).

Figure 4 Associations of KRT19 expression with TP53 and KRAS mutation status and their impact on overall survival. (A) Comparison of KRT19 expression between TP53 mutant (n=107) and wild-type tumors (n=69). KRT19 expression level was higher in TP53 mutant patients compared to wild-type patients (P<0.001). (B) Kaplan-Meier overall survival curves for patients with TP53 mutant versus those with wild-type. Patients harboring TP53 mutations had worse overall survival compared to wild-type patients (HR =1.667, 95% CI: 1.073–2.591, P=0.02). (C) Comparison of KRT19 expression between KRAS mutant (n=127) and wild-type tumors (n=49). KRT19 expression level was higher in KRAS mutant patients compared to wild-type patients (P<0.001). (D) Kaplan-Meier overall survival curves for patients with KRAS mutant versus those with wild-type. Patients harboring KRAS mutations had worse overall survival compared to wild-type patients (HR =1.807, 95% CI: 1.097–2.977, P=0.02). Data were derived from the TCGA-PAAD cohort. Differences in expression were analyzed by Student’s t-test, and survival differences were evaluated by the log-rank test. P values <0.05 were considered statistically significant. CI, confidence interval; HR, hazards ratio; KRT19, keratin 19; TCGA-PAAD, The Cancer Genome Atlas-Pancreatic Adenocarcinoma.

KRT19 expression is an independent prognostic factor for overall survival

Cox regression analysis revealed that in univariate analysis, high KRT19 expression (HR =1.446, P<0.001), high tumor grade (HR =1.498, P=0.009), and advanced AJCC stage (≥2b) (HR =2.079, P=0.006), TP53 mutation (HR =1.667, P=0.02) and KRAS mutation (HR =1.807, P=0.02) were associated with worse survival. In multivariate analysis, KRT19 expression (HR =1.665, P=0.001), and AJCC stage (≥2b) (HR =2.076, P=0.03) remained an independent prognostic factor (Table 2).

Discussion

In this study, the prognostic significance of KRT19 expression was comprehensively analyzed in PDAC using publicly available datasets and bioinformatics analyses. Our analysis revealed that KRT19 expression was markedly higher in tumor tissues than in normal pancreatic tissues and was significantly associated with a higher histological tumor grade. Patients with higher KRT19 expression exhibited poorer overall survival, and this association remained significant in the multivariate analysis, indicating that KRT19 is an independent prognostic factor. Moreover, high KRT19 expression positively correlated with TP53 and KRAS mutations, that are associated with worse clinical outcomes. These findings suggest that KRT19 plays a substantial role in tumor aggressiveness and KRT19 could be a potential indicator of poor prognosis.

A recent study suggested that KRT19 promotes PDAC progression by activating the hedgehog signaling pathway, which plays a key role in pancreatic cancer tumorigenesis and progression, and promotes the proliferation and metastasis of pancreatic cancer (8). Another study revealed that DNA methylation-associated allelic inactivation regulates KRT19 expression during pancreatic development. This study showed that DNA methylation is regulated by KRAS mutations and may promote carcinogenesis (15). Another study revealed that KRT19 is regulated by a specific miRNA, miR-374b-5p, that inhibits the Wnt/β-catenin pathway activation. This study also showed that hypoxic conditions suppressed miR-374b-5p expression and promoted KRT19 expression, eventually leading to pancreatic cancer progression (16).

KRT19 has been recognized as a biomarker associated with cancer metastasis, invasiveness, and poor prognosis among various malignancies (10,11). Similarly, previous studies have demonstrated its association with unfavorable clinical outcomes in patients with pancreatic cancer. However, most of these prior studies were single-institutional, limited by small sample sizes and relied mainly on univariate survival analyses, or inclusion of only limited clinical information, which hindered a comprehensive evaluation of KRT19 as a prognostic biomarker (2,4,5,9,17-19). In contrast, our study utilized a large-scale multi-institutional cohort derived from publicly available datasets. Notably, a very recent study also reported the association between KRT19 expression and TP53 status in PDAC (20). While their findings align with our observations regarding the prognostic significance of KRT19, our study provides further depth by performing comprehensive multivariate analyses. We demonstrated that KRT19 serves as an independent prognostic factor even after adjusting for key driver mutations, including KRAS and TP53. By integrating these molecular characteristics with survival data, our study offers a broader perspective on the biological relevance of KRT19, reinforcing its potential as a robust biomarker for predicting outcomes in PDAC patients.

Although our study provides novel evidence that KRT19 is an independent predictor for overall survival in patients with PDAC based on multivariate analysis, some limitations must be noted. First, the sample size is relatively small. Larger and more diverse patient cohorts are required to validate our findings and to strengthen the statistical power of multivariate studies. Second, as this study was based on a retrospective analysis of publicly available datasets, inherent selection bias and unmeasured confounding factors could not be excluded. Third, detailed treatment-related information, such as the use of adjuvant chemotherapy, radiotherapy, or surgical margin status, was not available in the datasets, which may have influenced the patient outcomes. Fourth, in this study, we focused on the individual prognostic value of KRT19 rather than a multi-gene panel. While PDAC is a highly heterogeneous disease driven by a complex interplay of numerous genetic alterations, the identification of a single, robust biomarker remains clinically significant for its potential as a straightforward diagnostic and prognostic tool in routine clinical practice. However, we acknowledge that a single-gene approach has inherent limitations in fully capturing the multi-layered molecular landscape of PDAC.

Despite these limitations, our findings underscore the potential role of KRT19 as an independent prognostic indicator in PDAC. Further studies are warranted to validate its utility as a biomarker, clarify the molecular mechanisms underlying its expression, and evaluate therapeutic strategies directed at KRT19. Such efforts may contribute to the development of targeted therapies and ultimately improve the outcomes of patients with PDAC.

Conclusions

Elevated KRT19 expression is associated with poor prognosis in PDAC and is correlated with more aggressive tumor features, including a higher tumor grade and the presence of TP53 and KRAS mutations. These findings suggest that KRT19 could serve as a valuable prognostic biomarker and potential therapeutic target in PDAC.

Acknowledgments

We would like to thank the contributors of the Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) projects for providing open access to their valuable datasets.

Footnote

Reporting Checklist: The authors have completed the REMARK reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2629/rc

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2629/prf

Funding: This study was supported by the Bisa Research Grant of Keimyung University in 2024 (No. 20240646).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1-2629/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Sarantis P, Koustas E, Papadimitropoulou A, et al. Pancreatic ductal adenocarcinoma: Treatment hurdles, tumor microenvironment and immunotherapy. World J Gastrointest Oncol 2020;12:173-81. [Crossref] [PubMed]
2. Wang Y, Liang C, Liu X, et al. A novel tumor-derived exosomal gene signature predicts prognosis in patients with pancreatic cancer. Transl Cancer Res 2024;13:4324-40. [Crossref] [PubMed]
3. Yu J, Ploner A, Kordes M, et al. Plasma protein biomarkers for early detection of pancreatic ductal adenocarcinoma. Int J Cancer 2021;148:2048-58. [Crossref] [PubMed]
4. Yao H, Yang Z, Liu Z, et al. Glypican-3 and KRT19 are markers associating with metastasis and poor prognosis of pancreatic ductal adenocarcinoma. Cancer Biomark 2016;17:397-404. [Crossref] [PubMed]
5. Li W, Li T, Sun C, et al. Identification and prognostic analysis of biomarkers to predict the progression of pancreatic cancer patients. Mol Med 2022;28:43. [Crossref] [PubMed]
6. Liu L, Wang S, Cen C, et al. Identification of differentially expressed genes in pancreatic ductal adenocarcinoma and normal pancreatic tissues based on microarray datasets. Mol Med Rep 2019;20:1901-14. [Crossref] [PubMed]
7. Moresco P, Kastan JP, Yang JI, et al. Signal peptide-independent secretion of keratin-19 by pancreatic cancer cells. Proc Natl Acad Sci U S A 2025;122:e2426218122. [Crossref] [PubMed]
8. Zhou C, Xiang Y, Ren Y, et al. Keratin19 promotes pancreatic cancer progression and poor prognosis via activating the Hedgehog pathway. Int J Oncol 2023;62:43. [Crossref] [PubMed]
9. Tang J, Zhuo H, Zhang X, et al. A novel biomarker Linc00974 interacting with KRT19 promotes proliferation and metastasis in hepatocellular carcinoma. Cell Death Dis 2014;5:e1549. [Crossref] [PubMed]
10. Kim H, Choi GH, Na DC, et al. Human hepatocellular carcinomas with "Stemness"-related marker expression: keratin 19 expression and a poor prognosis. Hepatology 2011;54:1707-17. [Crossref] [PubMed]
11. Kawai T, Yasuchika K, Ishii T, et al. Keratin 19, a Cancer Stem Cell Marker in Human Hepatocellular Carcinoma. Clin Cancer Res 2015;21:3081-91. [Crossref] [PubMed]
12. Lu XA, Pan YN, Xie CJ, et al. UGT1A6 in pancreatic cancer: a promising prognostic biomarker and therapeutic target. Clin Transl Oncol 2025;27:4450-65. [Crossref] [PubMed]
13. Mehrpouya M, Pourhashem Z, Yardehnavi N, et al. Evaluation of cytokeratin 19 as a prognostic tumoral and metastatic marker with focus on improved detection methods. J Cell Physiol 2019;234:21425-35. [Crossref] [PubMed]
14. Shang M, Zhang L, Chen X, et al. Identification of hub genes and regulators associated with pancreatic ductal adenocarcinoma based on integrated gene expression profile analysis. Discov Med 2019;28:159-72. [PubMed]
15. Krüger J, Fischer A, Breunig M, et al. DNA methylation-associated allelic inactivation regulates Keratin 19 gene expression during pancreatic development and carcinogenesis. J Pathol 2023;261:139-55. [Crossref] [PubMed]
16. Shi HQ, Li X, Chen Z, et al. KRT19 is regulated by miR-642a-5p and promotes pancreatic cancer progression through the Wnt/β-catenin pathway. iScience 2024;27:110782. [Crossref] [PubMed]
17. Saha SK, Choi HY, Kim BW, et al. KRT19 directly interacts with β-catenin/RAC1 complex to regulate NUMB-dependent NOTCH signaling pathway and breast cancer properties. Oncogene 2017;36:332-49. [Crossref] [PubMed]
18. Schmitt AM, Anlauf M, Rousson V, et al. WHO 2004 criteria and CK19 are reliable prognostic markers in pancreatic endocrine tumors. Am J Surg Pathol 2007;31:1677-82. [Crossref] [PubMed]
19. Jain R, Fischer S, Serra S, et al. The use of Cytokeratin 19 (CK19) immunohistochemistry in lesions of the pancreas, gastrointestinal tract, and liver. Appl Immunohistochem Mol Morphol 2010;18:9-15. [Crossref] [PubMed]
20. He L, Li L, Chen Z. Keratin 19 is a potential diagnostic and prognostic biomarker in pancreatic adenocarcinoma. Medicine (Baltimore) 2025;104:e46233. [Crossref] [PubMed]